View Single Post
  #20   Report Post  
Posted to sci.electronics.repair
[email protected] ohg...@gmail.com is offline
external usenet poster
 
Posts: 96
Default Finding the cmos battery

On Friday, January 8, 2021 at 12:55:24 PM UTC-5, Cursitor Doom wrote:
On Fri, 08 Jan 2021 11:44:55 -0600, Chuck
wrote:

On Thu, 07 Jan 2021 23:35:36 +0000, Cursitor Doom
wrote:

On Thu, 07 Jan 2021 13:33:31 -0500, legg wrote:

On Wed, 06 Jan 2021 20:51:28 +0000, Cursitor Doom
wrote:

On Wed, 06 Jan 2021 13:44:19 -0600, Chuck
wrote:

On Wed, 06 Jan 2021 18:00:32 +0000, Cursitor Doom
wrote:

Hi all,

I have a Roberts Stream 105 internet radio which I bought about 9
years ago. It's been fine up until maybe 18 months ago when it could
no longer remember my preferred settings. I'm guessing it has a little
backup cell in there somewhere that's gone way past its sell-by date.
I've opened up the case and it must be very well hidden indeed. I
asked Roberts for info on where it is but they obfuscated and told me
to return it to them and they'd fix it for 40 quid! There are no
manuals for this model on line (not service manuals anyway) and I *do*
like to fix things for myself as a matter of course anyway. The inside
consists of only 3 boards apart from the display: an audio board (as
it describes itself) the wireless card (I deduce from the fact that
although it's fully screened it's got a MAC address label on it) and
controller board interfacing to the user controls.
The only place this battery could be hiding is within a screened
enclosure on the "audio board". Now, I should have done this sooner
obviously, but time shortages and whatnot, I've probed the underside
of the screened area and found a persistent 0.3V above ground on some
of the joints. Does that sound like the sort of voltage a backup cell
would fall back to after 9 years? This 0.3V is with all external power
removed and after shorting out any capacitances.
Look for a 1/2 to1 farad electrolytic capacitor. Probably near the
upc.

At that value it would be a supercapacitor I'd imagine. Were they
installing those in new equipment 10 years ago?

They were showing up as surplus 20 years ago.

RL


Well I can't understand in that case why there are so few of them
listed on Ebay currently. I may have to imrovise here and go back to
my initial idea of using a lithium button cell in series with a diode
to prevent it being charged.

Anyone see any issues with that approach?

Here is one that is genuine and exactly like the ones I used to use as
a replacement.
https://www.newark.com/kemet/ft0h474...al/dp/78AC5690

Thanks, but you guys have avoided answering the question about
replacing the supercap with a cell of whatever capacity in series with
a low reverse-leakage diode. Anyone know what the likely voltage that
cell would need to be?


Generally speaking, I've found that supercaps used as memory devices are generally run *very* close to their voltage rating, so if your supercap is a 2.7V, there's a good chance there could be 2.5 volts on it. But, you don't need to make this complicated. Plug the radio in and see what the voltage is across the supercap. Most of the supercaps I've changed just go low value/high ESR and don't short. Whatever the voltage is under test, add that via cell. Since the value you need is unlikely (Murphy's Law) to be a direct value of a cell you want to add, you may have to add more than one diode to drop the voltage to or even a bit below the memory voltage.